4.5 Article

Nonconforming Schwarz-spectral element methods for incompressible flow

Journal

COMPUTERS & FLUIDS
Volume 191, Issue -, Pages -

Publisher

PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.compfluid.2019.104237

Keywords

Overlapping-Schwarz; Overset; High-order; Fluid-dynamics; Heat-transfer; Scalability

Funding

  1. U.S. Department of Energy, Office of Science, the Office of Advanced Scientific Computing Research [DE-AC02-06CH11357]
  2. National Science Foundation [OCI-0725070, ACI-1238993]
  3. state of Illinois

Ask authors/readers for more resources

We present scalable implementations of spectral-element-based Schwarz overlapping (overset) methods for the incompressible Navier-Stokes (NS) equations. Our SEM-based overset grid method is implemented at the level of the NS equations, which are advanced independently within separate subdomains using interdomain boundary-data exchanges at each timestep or sub-timestep. Central to this implementation is a general, robust, and scalable interpolation routine, that rapidly determines the computational coordinates for arbitrary points x* = (x*, y*, z*) is an element of Omega subset of R-3. The communication kernels in gslib execute with at most log P complexity for P MPI ranks and have scaled to P> 10(6). Given their performance and robustness, they obviate the need for development of additional MPI-based code for the Schwarz implementation and thus greatly simplify the development of a scalable parallel Schwarz solver. The communication overhead due to the boundary-data interpolation and exchange is only about 1% of the total time-tosolution for most cases. The original interpolation routine has been extended to support integer and real discriminator fields to choose which domain is responsible for interpolation when more than two subdomains overlap in a given region. We discuss the computation/communication complexity and accuracy of the approach, and present performance measurements for P> 12, 000 processors. We also demonstrate convergence results for the Schwarz-SEM formulation in multiple 2D and 3D configurations and present application of this method to several challenging fluid and heat transfer problems. (C) 2019 Published by Elsevier Ltd.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.5
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

Article Nuclear Science & Technology

DEPENDENCE OF FIRE TIME OF CONCERN ON LOCATION OF A ONE-ASSEMBLY TRANSPORT PACKAGE

Ketan Mittal, Ahti Suo-Anttila, Miles Greiner

NUCLEAR TECHNOLOGY (2015)

Article Mathematics, Applied

Mesh Smoothing for the Spectral Element Method

Ketan Mittal, Paul Fischer

JOURNAL OF SCIENTIFIC COMPUTING (2019)

Article Mathematics, Applied

THE TARGET-MATRIX OPTIMIZATION PARADIGM FOR HIGH-ORDER MESHES

Veselin Dobrev, Patrick Knup, Tzanio Kolev, Ketan Mittal, Vladimir Tomov

SIAM JOURNAL ON SCIENTIFIC COMPUTING (2019)

Article Computer Science, Interdisciplinary Applications

Simulation-driven optimization of high-order meshes in ALE hydrodynamics

Veselin Dobrev, Patrick Knupp, Tzanio Kolev, Ketan Mittal, Robert Rieben, Vladimir Tomov

COMPUTERS & FLUIDS (2020)

Article Computer Science, Interdisciplinary Applications

Direct numerical simulation of rotating ellipsoidal particles using moving nonconforming Schwarz-spectral element method

Ketan Mittal, Som Dutta, Paul Fischer

COMPUTERS & FLUIDS (2020)

Article Mechanics

Direct numerical simulation of the turbulent flow generated during a violent expiratory event

Alexandre Fabregat, Ferran Gisbert, Anton Vernet, Som Dutta, Ketan Mittal, Jordi Pallares

Summary: This study numerically investigates the hydrodynamics produced by a violent expiratory event resembling a mild cough, showing that factors like exhaled velocity, buoyancy, and temperature differences affect gas dispersion dynamics. Numerical results suggest that while analytical models provide reasonable estimates of the distance traveled by the gas puff, trajectory predictions show larger deviations from the direct numerical simulation (DNS).

PHYSICS OF FLUIDS (2021)

Article Computer Science, Interdisciplinary Applications

hr-Adaptivity for nonconforming high-order meshes with the target matrix optimization paradigm

Veselin Dobrev, Patrick Knupp, Tzanio Kolev, Ketan Mittal, Vladimir Tomov

Summary: The study introduces an hr-adaptivity framework for optimization of high-order meshes, extending the r-adaptivity method with nonconforming adaptive mesh refinement to better satisfy geometric targets. The methodology is purely algebraic, applicable to various types of meshes and dimensions, and achieves similar accuracy results with significantly fewer mesh nodes.

ENGINEERING WITH COMPUTERS (2022)

Article Mechanics

Direct numerical simulation of turbulent dispersion of evaporative aerosol clouds produced by an intense expiratory event

Alexandre Fabregat, Ferran Gisbert, Anton Vernet, Josep Anton Ferre, Ketan Mittal, Som Dutta, Jordi Pallares

Summary: Airborne particles play a significant role in the transmission of COVID-19 and other infectious diseases, particularly when individuals talk, sing, cough, or sneeze. The dynamics of particles dispersed in the air evolve differently based on their sizes, with specific aerosol diameters showing increased residence time and horizontal range under typical ambient conditions.

PHYSICS OF FLUIDS (2021)

Article Computer Science, Interdisciplinary Applications

Multirate timestepping for the incompressible Navier-Stokes equations in overlapping grids

Ketan Mittal, Som Dutta, Paul Fischer

Summary: The study introduces a multirate timestepper for semi-implicit solutions of the unsteady incompressible Navier-Stokes equations, based on a multidomain spectral element method. The novelty lies in the development of a stable overlapping Schwarz method directly applied to the Navier-Stokes equations, rather than to the convective, viscous, and pressure substeps commonly used in most solvers.

JOURNAL OF COMPUTATIONAL PHYSICS (2021)

Article Mechanics

Impact of gaps on the flow statistics in an emergent rigid canopy

Pallav Ranjan, Ketan Mittal, Leonardo P. Chamorro, Rafael O. Tinoco

Summary: High-resolution large eddy simulations and laboratory experiments were used to evaluate the flow response to gaps in cylinder arrays. The results showed that the gaps affected the flow statistics in the upstream and downstream regions of the canopy, and the in-plane turbulent kinetic energy exhibited a consistent decay rate. The emergent canopy acted as a passive turbulence generator for the gap flow.

PHYSICS OF FLUIDS (2022)

Article Multidisciplinary Sciences

Multi-fidelity modelling of shark skin denticle flows: insights into drag generation mechanisms

C. J. Lloyd, K. Mittal, S. Dutta, R. M. Dorrell, J. Peakall, G. M. Keevil, A. D. Burns

Summary: This study investigates the flow over smooth denticles on shark skin in an open-channel flow. It is found that there are large pressure peaks and viscous drag at the edges of the denticles. Although the denticles produce lift forces and positive spanwise torques, they ultimately increase the drag by 58% compared to a flat plate. RANS models provide good predictions of drag distributions, although they underestimate the turbulent kinetic energy production and subsequently the drag.

ROYAL SOCIETY OPEN SCIENCE (2023)

Article Computer Science, Software Engineering

High-Order Mesh Morphing for Boundary and Interface Fitting to Implicit Geometries

Jorge-Luis Barrera, Tzanio Kolev, Ketan Mittal, Vladimir Tomov

Summary: We propose a method for morphing high-order meshes to align with implicitly defined geometries. The method formulates the mesh optimization problem using a variational minimization approach and a penalty term. It utilizes a source mesh to represent the level set function accurately and incorporates adaptive strategies for setting the penalization weight and selecting the faces of the mesh to fit the target isocontour. The proposed method is demonstrated to be robust for generating boundary-and interface-fitted meshes for curvilinear domains.

COMPUTER-AIDED DESIGN (2023)

Proceedings Paper Engineering, Mechanical

THERMAL ANALYSIS OF A NAC-LWT CASK UNDER NORMAL AND FIRE ACCIDENT CONDITIONS

Ketan Mittal, Miles Greiner

PROCEEDINGS OF THE ASME PRESSURE VESSELS AND PIPING CONFERENCE, PVP 2012, VOL 7: OPERATIONS, APPLICATIONS AND COMPONENTS (2012)

Article Computer Science, Interdisciplinary Applications

Well-balanced kinetic schemes for two-phase flows

Jin Bao, Zhaoli Guo

Summary: At the equilibrium state of a two-phase fluid system, the chemical potential is constant and the velocity is zero. However, it is challenging to capture this equilibrium state accurately in numerical simulations, resulting in inconsistent thermodynamic interfacial properties and spurious velocities. Therefore, numerical schemes with well-balanced properties are preferred for simulating two-phase flows.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

High-Order Implicit Large Eddy Simulation using Entropically Damped Artificial Compressibility

Brian C. Vermeire

Summary: This study presents a framework for implicit large eddy simulation (ILES) of incompressible flows by combining the entropically damped artificial compressibility (EDAC) method with the flux reconstruction (FR) approach. Experimental results demonstrate that the method is accurate and stable for low-order solutions, while higher-order solutions exhibit significantly higher accuracy and lower divergence error compared to reference direct numerical simulation.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Analysis of inflow conditions on the flow past a wall-mounted square cylinder with OpenFOAM

Mijian Li, Rui Wang, Xinyu Guo, Xinyu Liu, Lianzhou Wang

Summary: In this study, the flow mechanisms around wall-mounted structures were investigated using Large Eddy Simulation (LES). The impact of inflow turbulence on the flow physics, dynamic response, and hydrodynamic performance was explored. The results revealed strong interference between velocity fluctuations and the wake past the cylinder, as well as significant convection effects in the far wake region.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Evaluation of a high-order central-difference solver for highly compressible flows out of thermochemical equilibrium

Donatella Passiatore, Luca Sciacovelli, Paola Cinnella, Giuseppe Pascazio

Summary: A high-order shock-capturing central finite-difference scheme is evaluated for numerical simulations of hyper-sonic high-enthalpy flows out of thermochemical equilibrium. The scheme utilizes a tenth-order accurate central-difference approximation of inviscid fluxes, along with high-order artificial dissipation and shock-capturing terms. The proposed approach demonstrates accuracy and robustness for a variety of thermochemical non-equilibrium configurations.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Sessile super-droplets as quasi-static wall deformations in direct numerical simulation of turbulent channel flow

Philipp Bahavar, Claus Wagner

Summary: Condensation is an important aspect in flow applications, and simulating the gas phase and tracking the deposition rates of condensate droplets can capture the effects of surface droplets on the flow while reducing computational costs.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Reused LU factorization as a preconditioner for efficient solution of the parabolized stability equations

Andras Szabo, Gyorgy Paal

Summary: This paper introduces an efficient calculation method, the parabolized stability equations (PSE), for solving stability equations. By calculating LU factorization once in each marching step, the time spent on solving linear systems of equations can be significantly reduced. Numerical experiments demonstrate the effectiveness of this method in reducing the solution time for linear equations, and its applicability to similar problems.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Evaluation of an efficient data-driven ANN model to predict agglomerate collisions within Euler-Lagrange simulations

A. Khalifa, M. Breuer

Summary: This study evaluates a recently developed data-driven model for collision-induced agglomerate breakup in high mass loading flows. The model uses artificial neural networks to predict the post-collision behavior of agglomerates, reducing computational costs compared to coupled CFD-DEM simulations.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

A high-order domain preserving DG method for the two-layer shallow water equations

Chunmei Du, Maojun Li

Summary: This paper considers the bilayer shallow water wave equations in one-dimensional space and presents an invariant domain preserving DG method to avoid Kelvin-Helmholtz instability.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Learning of viscosity functions in rarefied gas flows with physics-informed neural networks

Jean-Michel Tucny, Mihir Durve, Andrea Montessori, Sauro Succi

Summary: The prediction of non-equilibrium transport phenomena in disordered media is a challenging problem for conventional numerical methods. Physics-informed neural networks (PINNs) show potential for solving this inverse problem. In this study, PINNs were used to successfully predict the velocity field of rarefied gas flow, and AdamW was found to be the best optimizer.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Shock capturing for a high-order ALE discontinuous Galerkin method with applications to fluid flows in time-dependent domains

Min Gao, Pascal Mossier, Claus-Dieter Munz

Summary: In recent decades, the arbitrary Lagrangian-Eulerian (ALE) approach has gained popularity in dealing with fluid flows with moving boundaries. This paper presents a novel algorithm that combines the ALE finite volume (FV) and ALE discontinuous Galerkin (DG) methods into a stable and efficient hybrid approach. The main challenge of this mixed ALE FV and ALE DG method is reducing the inconsistency between the two discretizations. The proposed algorithm is implemented into a loosely-coupled fluid-structure interaction (FSI) framework and is demonstrated through various benchmark test cases and complex scenarios.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Study of the convergence of the Meshless Lattice Boltzmann Method in Taylor-Green, annular channel and a porous medium flows

Dawid Strzelczyk, Maciej Matyka

Summary: In this study, the numerical convergence of the Meshless Lattice Boltzmann Method (MLBM) is investigated through three benchmark tests. The results are compared to the standard Lattice Boltzmann Method (LBM) and the analytical solution of the Navier-Stokes equation. It is found that MLBM outperforms LBM in terms of error value for the same number of nodes discretizing the domain.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

An adaptive approach to remove tensile instability in SPH for weakly compressible fluids

Kanishka Bhattacharya, Tapan Jana, Amit Shaw, L. S. Ramachandra, Vishal Mehra

Summary: In this work, an adaptive algorithm is developed to address the issue of tensile instability in Smoothed Particle Hydrodynamics (SPH) by adjusting the shape of the kernel function to satisfy stability conditions. The effectiveness of the algorithm is demonstrated through dispersion analysis and fluid dynamics simulations.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Analysis of improved digital filter inflow generation methods for compressible turbulent boundary layers

Luis Laguarda, Stefan Hickel

Summary: We propose several enhancements to improve the accuracy and performance of the digital filter turbulent inflow generation technique, such as introducing a more realistic correlation function and varying target length scales. Additionally, we suggest generating inflow data in parallel at a prescribed time interval to improve computational performance. Based on the results of large-eddy simulations, these enhancements have shown to be beneficial. Suppressing streamwise velocity fluctuations at the inflow leads to the fastest relaxation of pressure fluctuations. However, this approach increases the adaptation length, which can be shortened by artificially increasing the wall-normal Reynolds stresses.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

A compressible multiphase Mass-of-Fluid model for the simulation of laser-based manufacturing processes

Constantin Zenz, Michele Buttazzoni, Tobias Florian, Katherine Elizabeth Crespo Armijos, Rodrigo Gomez Vazquez, Gerhard Liedl, Andreas Otto

Summary: A new model for compressible multiphase flows involving sharp interfaces and phase change is presented, with a focus on the treatment of compressibility and phase change in the multiphase fluid flow model. The model's accuracy and suitability are demonstrated through comparisons with experimental observations.

COMPUTERS & FLUIDS (2024)

Article Computer Science, Interdisciplinary Applications

Quantifying uncertainties in direct numerical simulations of a turbulent channel flow

Joseph O'Connor, Sylvain Laizet, Andrew Wynn, Wouter Edeling, Peter V. Coveney

Summary: This article aims to apply uncertainty quantification and sensitivity analysis to the direct numerical simulation (DNS) of low Reynolds number wall-bounded turbulent channel flow. By using a highly scalable DNS framework and UQ techniques, the study evaluates the influence of different numerical parameters on the simulation results without explicitly modifying the code. The findings provide guidance for numerical simulations of wall-bounded turbulent flows.

COMPUTERS & FLUIDS (2024)